1. Read Section 1.4, Section 1.7 (pp. 48-49)
Performance
Read Section 1.4, Section 1.7 (pp )
Adapted from Slides by Prof. Mary Jane Irwin, Penn State University
And Slides Supplied by the Publisher

2. Outline What is computer architecture? Classes of Computers
Performance
Performance Metrics
The performance equation
Measuring performance
Improving performance: parallelism, locality, Amdahl's law

3. What is Computer Architecture?
Old definition of computer architecture = Instruction set architecture, ISA

4. Instruction set architecture, ISA
Organization of Programmable Storage
Data Types & Data Structures: Encodings & Representations
Instruction Formats
Instruction Set
Modes of Addressing and Accessing Data Items and Instructions
How are “Exceptional Conditions” Handled

5. ISA vs. Computer Architecture
Architect’s job much more than instruction set design
technical hurdles today more challenging than those in instruction set design

6. Definition of Computer Architecture
The science and art of selecting and interconnecting hardware components to create computers that meet functional, performance and cost goals.
Computer architecture >> ISA

7. Classes of Computers Desktop computers Servers Supercomputers
Designed to deliver good performance to a single user at low cost usually executing 3rd party software, usually incorporating a graphics display, a keyboard, and a mouse
Servers
Used to run larger programs for multiple, simultaneous users typically accessed only via a network and that places a greater emphasis on dependability and security
Supercomputers
A high performance, high cost class of servers with hundreds to thousands of processors, terabytes (240 ≈1012 ) of memory and petabytes (250 ≈1015 ) of storage that are used for high-end scientific and engineering applications
Embedded computers (processors)
A computer inside another device used for running one predetermined application

8. Review: Some Basic Definitions
Kilobyte – 210 or 1,024 bytes
Megabyte– 220 or 1,048,576 bytes
sometimes “rounded” to 106 or 1,000,000 bytes
Gigabyte – 230 or 1,073,741,824 bytes
sometimes rounded to 109 or 1,000,000,000 bytes
Terabyte – 240 or 1,099,511,627,776 bytes
sometimes rounded to 1012 or 1,000,000,000,000 bytes
Petabyte – 250 or 1024 terabytes
sometimes rounded to 1015 or 1,000,000,000,000,000 bytes
Exabyte – 260 or 1024 petabytes
Sometimes rounded to 1018 or 1,000,000,000,000,000,000 bytes

9. Outline What is computer architecture? Performance Performance metrics
The performance equation
Measuring performance
Improving performance: parallelism, locality, Amdahl's law

10. Performance Metrics Purchasing perspective Design perspective
given a collection of computers, which has the
best performance ?
least cost ?
best cost/performance?
Design perspective
faced with design options, which has the
best performance improvement ?
Both require
basis for comparison
metric for evaluation
Or smallest/lightest
Longest battery life
Most reliable/durable (in space)

11. Response Time versus Throughput
Response time (execution time):
the time between the start and the completion of a task
Important to individual users
Throughput (bandwidth) – the total amount of work done in a given time
Important to computer center managers
Will need different performance metrics as well as a different set of applications to benchmark embedded and desktop computers, which are more focused on response time, versus servers, which are more focused on throughput

12. Defining (Speed) Performance
To maximize performance, we need to minimize execution time
Consider a computer X
performanceX = 1 / execution_timeX
If computer X is n times faster than computer Y, then
performanceX execution_timeY
= = n
performanceY execution_timeX
Increasing performance requires decreasing execution time
Decreasing response time almost always improves throughput

13. Relative Performance Example
If computer A runs a program in 10 seconds and computer B runs the same program in 15 seconds, how much faster is A than B?
We know that A is n times faster than B if
performanceA execution_timeB
= = n
performanceB execution_timeA
The performance ratio is 15
= 1.5 10
So A is 1.5 times faster than B

14. Performance Factors CPU execution time of a program (CPU time):
time the CPU spends running the program
Does not include time waiting for Input or output (I/O) operations or running other programs or
Can improve performance by reducing either the length of the clock cycle or the number of clock cycles required for a program or both.
or the number of clock cycles required for a program or both.

15. Review: Machine Clock Rate
Clock rate (clock cycles per second in MHz or GHz) is inverse of clock cycle time (clock period)
CC = 1 / CR
one clock period
10 nsec clock cycle => 100 MHz clock rate
5 nsec clock cycle => 200 MHz clock rate
2 nsec clock cycle => 500 MHz clock rate
1 nsec (10-9) clock cycle => 1 GHz (109) clock rate
500 psec clock cycle => 2 GHz clock rate
250 psec clock cycle => 4 GHz clock rate
200 psec clock cycle => 5 GHz clock rate
A clock cycle is the basic unit of time to execute one operation/pipeline stage/etc.

16. Improving Performance Example
A program runs on computer A with a 2 GHz clock in 10 seconds.
What clock rate must computer B have to run this program in 6 seconds?
Note: Computer B will require 1.2 times as many clock cycles as computer A to run the program.
For lecture

17. Improving Performance Example
A program runs on computer A with a 2 GHz clock in 10 seconds.
What clock rate must computer B have to run this program in 6 seconds?
Note: Computer B will require 1.2 times as many clock cycles as computer A to run the program.
CPU timeA CPU clock cyclesA
clock rateA =
For lecture
CPU clock cyclesA = 10 sec x 2 x 109 cycles/sec = 20 x 109 cycles
CPU timeB x 20 x 109 cycles
clock rateB =
clock rateB x 20 x 109 cycles
6 seconds
= = 4 GHz

18. Clock Cycles Per Instruction, CPI
Not all instructions take the same amount of time to execute
A program execution time is equals the number of instructions executed multiplied by the average time per instruction
Clock cycles per instruction (CPI) – the average number of clock cycles each instruction takes to execute
CPI for this instruction class A B C
CPI 1 2 3

19. Using the Performance Equation
Computers A and B implement the same ISA.
Computer A has a clock cycle time of 250 ps and an effective CPI of 2.0 for some program
Computer B has a clock cycle time of 500 ps and an effective CPI of 1.2 for the same program.
Which computer is faster and by how much?
For class handout

20. Using the Performance Equation
Computer A clock cycle time = 250 ps ; CPIA = 2.0
Computer B clock cycle time = 500 ps ; CPIB = 1.2
Which computer is faster and by how much?
Both A and B execute the same number of instructions, I, so
CPU timeA = I x 2.0 x 250 ps = 500 x I ps
CPU timeB = I x 1.2 x 500 ps = 600 x I ps
For lecture
Clearly, A is faster … by the ratio of execution times
Clearly, A is faster … by the ratio of execution times
performanceA execution_timeB x I ps
= = = 1.2
performanceB execution_timeA x I ps

21. Morgan Kaufmann Publishers
April 14, 2017
CPI in More Detail
If n different instruction types (classes) are executed by a program
Each instruction of type i takes a different number of cycles to execute (CPIi)
Hence, the average # of clocks per instruction, CPI
Relative frequency of instruction type i
Dr. W. Abu-Sufah 21
Chapter 1 — Computer Abstractions and Technology

22. Effective (Average) CPI
Hence, computing the overall effective CPI is done by looking at the different types (classes) of instructions executed by the program and their individual cycle counts and then averaging
Let n be the number of different instruction classes executed by a program
Freqi is the relative frequency {percentage} of class i instructions executed
CPIi is the average number of clock cycles per instruction for instruction class i
Then we have:
Overall effective CPI =  (CPIi x Freqi)
i = 1 n
The overall effective CPI varies by instruction mix – a measure of the dynamic frequency of instructions across one or many programs

23. THE Performance Equation
Our basic performance equation is then
CPU time = Instruction_count x CPI x clock_cycle or
Instruction_count x CPI
clock_rate
CPU time =
These equations separate the three key factors that affect performance: Instruction_coun, CPI, and clock_rate
Can measure the CPU execution time by running the program
The clock rate is usually given
Can measure overall instruction count by using profilers/ simulators without knowing all of the implementation details
CPI varies by instruction type and ISA implementation for which we must know the implementation details
Note that instruction count is dynamic – i.e., its not the number of lines in the code, but THE NUMBER OF INSTRUCTIONS EXECUTED

24. Determinates of CPU Performance
CPU time = Instruction_count x CPI x clock_cycle
Instruction _count
CPI
clock_cycl e
Algorithm
Programming language
Compiler
ISA
Core organization
Technology
For class handout

25. Determinates of CPU Performance
CPU time = Instruction_count x CPI x clock_cycle
Instruction _count
CPI
clock_cycle
Algorithm
Programming language
Compiler
ISA
Core organization
Technology X X X X X X
For lecture X X X X X X

26. A Simple Example Op
Freq
CPIi
Freq x CPIi
ALU
50% 1 .
Load
20% 5
Store
10% 3
Branch 2
 =
How much faster would the machine be if a better data cache reduced the average load time to 2 cycles?
How does this compare with using branch prediction to shave a cycle off the branch time?
What if two ALU instructions could be executed at once?

27. A Simple Example Op
Freq
CPIi
Freq x CPIi
ALU
50% 1
Load
20% 5
Store
10% 3
Branch 2
 = .5
1.0 .3 .4 .5 .4 .3 .5
1.0 .3 .2
.25
1.0 .3 .4
2.2
1.6
2.0
1.95
How much faster would the machine be if a better data cache reduced the average load time to 2 cycles?
How does this compare with using branch prediction to shave a cycle off the branch time?
What if two ALU instructions could be executed at once?
For lecture
CPU time new = 1.6 x IC x CC so 2.2/1.6 means 37.5% faster
CPU time new = 2.0 x IC x CC so 2.2/2.0 means 10% faster
CPU time new = 1.95 x IC x CC so 2.2/1.95 means 12.8% faster

28. Before We Continue
Lecture notes are posted on the class web-site usually before the lecture.
Review them and try to get some understanding if possible before class.
Read assigned textbook sections
Homework1 has been posted since Friday, February 8. There are 10 problems to solve, several with many parts.
The homework is time consuming and can't be solved in one night. You should have started working on them by now.
Print your solution of the homework and bring it to class next Tuesday, February 19.
Do not cheat. Cheating will be pursued and it defeats the purpose of trying to solve the homework.

29. Workloads and Benchmarks
Benchmarks – a set of programs that form a “workload” specifically chosen to measure performance
SPEC (System Performance Evaluation Cooperative) creates standard sets of benchmarks starting with SPEC89. The latest is SPEC CPU2006 which consists of 12 integer benchmarks (CINT2006) and 17 floating- point benchmarks (CFP2006).
There are also benchmark collections for power workloads (SPECpower_ssj2008), for mail workloads (SPECmail2008), for multimedia workloads (mediabench), …

30. SPEC CINT2006 on Barcelona (CC = 0.4 x 109)
See Slide notes
Name
ICx109
CPI
ExTime
RefTime
SPEC ratio
perl
21,118
0.75
637
9,770
15.3
bzip2
2,389
0.85
817
9,650
11.8
gcc
1,050
1.72
724
8,050
11.1
mcf
336
10.00
1,345
9,120
6.8 go
1,658
1.09
721
10,490
14.6
hmmer
2,783
0.80
890
9,330
10.5
sjeng
2,176
0.96
837
12,100
14.5
libquantum
1,623
1.61
1,047
20,720
19.8
h264avc
3,102
993
22,130
22.3
omnetpp
587
2.94
690
6,250
9.1
astar
1,082
1.79
773
7,020
xalancbmk
1,058
2.70
1,143
6,900
6.0
Geometric Mean
11.7
Perl – interpreted string processing
Bzip2 – block-sorting compression
Gcc – GNU C compiler
Mcf – combinatorial optimization
Go – Go came (AI)
Hmmer – search gene sequence
Sjeng – chess game (AI)
Libquantum – quantum computer simulation
H264avc – video compression
Omnetpp – discret event simulation library
Astar – games/path finding
Xalancbmk – XML parsing
High cache miss rates

31. Comparing and Summarizing Performance
How do we summarize the performance for a benchmark set (e.g. SPEC CINT2006 ) with a single number?
First the execution times are normalized giving the “SPEC ratio” (bigger is faster, i.e., SPEC ratio is the inverse of execution time)
The SPEC ratios are then “averaged” using the geometric mean (GM)
GM = n  SPEC ratioi
i = 1 n

32. Comparing and Summarizing Performance (cont.)
Guiding principle in reporting performance measurements is reproducibility
List everything another experimenter would need to duplicate the experiment
version of the operating system
compiler settings
input set used
specific computer configuration (clock rate, cache sizes and speed, memory size and speed, etc.)

33. Summary: Evaluating ISAs
Design-time metrics:
Can the ISA be implemented, in how long, at what cost?
Can the ISA be programmed? Ease of compilation?
Static metrics:
How many bytes does the program occupy in memory?
Dynamic (run time) metrics:
How many instructions are executed? How many bytes does the processor fetch to execute the program?
How many clocks are required per instruction?
How small a clock cycle is practical?
Best Metric: Time to execute the program!
CPI
Inst. Count
Cycle Time
depends on the instructions set, the processor organization, and compilation techniques.

34. Morgan Kaufmann Publishers
April 14, 2017
Performance Summary
Performance depends on
Algorithm: affects IC, possibly CPI
Programming language: affects IC, CPI
Compiler: affects IC, CPI
Instruction set architecture: affects IC, CPI, Tclock
Chapter 1 — Computer Abstractions and Technology